INDEX

Explanations

expressions related to explaining or clarifying something

oai_token-act-pair · gpt-3.5-turbo

phrases indicating intentions or propositions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ADRA

-0.68

 watershed

-0.64

 Lieberman

-0.63

 vigilance

-0.62

meet

-0.60

 novelty

-0.60

rule

-0.59

Dub

-0.59

TIM

-0.59

 extrad

-0.58

POSITIVE LOGITS

orah

0.96

aucus

0.71

ften

0.70

hower

0.70

eous

0.70

hesive

0.69

soType

0.69

oice

0.68

ptoms

0.67

oke

0.67

Activations Density 0.082%

expressions related to explaining or clarifying something

phrases indicating intentions or propositions

No Comments

No Known Activations